edsl 0.1.31.dev4__py3-none-any.whl → 0.1.33__py3-none-any.whl

edsl/questions/prompt_templates/question_numerical.jinja

@@ -0,0 +1,37 @@
+You are being asked a question that requires a numerical response 
+in the form of an integer or decimal (e.g., -12, 0, 1, 2, 3.45, ...).
+        
+Your response must be in the following format:
+
+{% if include_comment %}
+    {"answer": "<your numerical answer here>", "comment": "<your explanation here>"}
+{% else %}
+    {"answer": "<your numerical answer here>"}
+{% endif %}
+
+You must only include an integer or decimal in the quoted "answer" part of your response.
+        
+Here is an example of a valid response:
+{% if include_comment %}
+    {"answer": "100", "comment": "This is my explanation..."}
+{% else %}
+    {"answer": "100"}
+{% endif %}
+        
+Here is an example of a response that is invalid because the "answer" includes words:
+{"answer": "I don't know.", ...}
+
+If your response is equivalent to zero, your formatted response should look like this:
+{% if include_comment %}
+    {"answer": "0", "comment": "This is my explanation..."}
+{% else %}
+    {"answer": "0"}
+{% endif %}
+
+You are being asked the following question: {{question_text}}
+{% if min_value is not none %}
+    Minimum answer value: {{min_value}}
+{% endif %}
+{% if max_value is not none %}
+    Maximum answer value: {{max_value}}
+{% endif %}

edsl/questions/question_registry.py

@@ -100,12 +100,16 @@ class Question(metaclass=Meta):
 
         >>> from edsl import Question
         >>> Question.available()
-        ['budget', 'checkbox', 'extract', 'free_text', 'functional', 'likert_five', 'linear_scale', 'list', 'multiple_choice', 'numerical', 'rank', 'top_k', 'yes_no']
+        ['checkbox', 'extract', 'free_text', 'functional', 'likert_five', 'linear_scale', 'list', 'multiple_choice', 'numerical', 'rank', 'top_k', 'yes_no']
         """
+        exclude = ["budget"]
         if show_class_names:
             return RegisterQuestionsMeta.question_types_to_classes()
         else:
-            return sorted(set(RegisterQuestionsMeta.question_types_to_classes().keys()))
+            question_list = sorted(
+                set(RegisterQuestionsMeta.question_types_to_classes().keys())
+            )
+            return [q for q in question_list if q not in exclude]
 
 
 def get_question_class(question_type):

edsl/questions/templates/budget/answering_instructions.jinja

@@ -0,0 +1,7 @@
+Return only a comma-separated list the values in the same order as the options, with 0s included, on one line, in square braces.
+
+Example: if there are 4 options, the response should be "[25,25,25,25]" to allocate 25 to each option.
+
+{% if include_comment %}
+After the answer, you can put a comment explaining your choice on the next line.
+{% endif %}

edsl/questions/templates/budget/question_presentation.jinja

@@ -0,0 +1,7 @@
+{{question_text}}
+The options are 
+{% for option in question_options %}
+{{ loop.index0 }}: {{option}}
+{% endfor %}                       
+Allocate your budget of {{budget_sum}} among the options. 
+

edsl/questions/templates/checkbox/answering_instructions.jinja

@@ -0,0 +1,10 @@
+{# Answering Instructions #}
+{% if use_code %}
+Please respond only with a comma-separated list of the code of the options that apply, with square brackets. E.g., [0, 1, 3]
+{% else %}
+Please respond only with a comma-separated list of the options that apply, with square brackets. E.g., ['Good', 'Bad', 'Ugly']
+{% endif %}
+{% if include_comment %}
+After the answer, you can put a comment explaining your choice on the next line.
+{% endif %}
+

edsl/questions/templates/checkbox/question_presentation.jinja

@@ -0,0 +1,22 @@
+{{question_text}}
+{% if use_code %}
+    {% for option in question_options %}
+{{ loop.index0 }}: {{option}}
+    {% endfor %}
+{% else %}
+    {% for option in question_options %}
+{{ option }}
+    {% endfor %}    
+{% endif %}
+
+{# Restrictions #}
+{% if min_selections != None and max_selections != None and min_selections == max_selections %}
+You must select exactly {{min_selections}} options.
+{% elif min_selections != None and max_selections != None %}
+Minimum number of options that must be selected: {{min_selections}}.
+Maximum number of options that must be selected: {{max_selections}}.
+{% elif min_selections != None %}
+Minimum number of options that must be selected: {{min_selections}}.
+{% elif max_selections != None %}
+Maximum number of options that must be selected: {{max_selections}}.
+{% endif %}

edsl/questions/templates/extract/answering_instructions.jinja

@@ -0,0 +1,7 @@
+An ANSWER should be formatted like this: 
+
+{{ answer_template }}
+
+It should have the same keys but values extracted from the input.
+If the value of a key is not present in the input, fill with "null".
+Put any comments in the next line after the answer.

edsl/questions/templates/extract/question_presentation.jinja

@@ -0,0 +1 @@
+{{question_text}}

edsl/questions/templates/free_text/question_presentation.jinja

@@ -0,0 +1 @@
+{{question_text}}

edsl/questions/templates/likert_five/answering_instructions.jinja

@@ -0,0 +1,10 @@
+{# Answering Instructions #}
+{% if use_code %}
+Respond only with the code corresponding to one of the options.
+{% else %}
+Respond only with a string corresponding to one of the options.
+{% endif %}
+{% if include_comment %}
+After the answer, you can put a comment explaining why you chose that option on the next line.
+{% endif %}
+

edsl/questions/templates/likert_five/question_presentation.jinja

@@ -0,0 +1,12 @@
+{# Question Presention #}
+{{question_text}}
+{% if use_code %}
+    {%- for option in question_options %}
+{{ loop.index0 }}: {{option}}
+    {% endfor %}
+{% else %}
+    {% for option in question_options %}
+{{option}}
+    {% endfor %}
+{% endif %}
+Only 1 option may be selected.

edsl/questions/templates/linear_scale/answering_instructions.jinja

@@ -0,0 +1,5 @@
+{# Answering Instructions #}
+Respond only with the code corresponding to one of the options. E.g., "1" or "5" by itself.
+{% if include_comment %}
+After the answer, you can put a comment explaining why you chose that option on the next line.
+{% endif %}

edsl/questions/templates/linear_scale/question_presentation.jinja

@@ -0,0 +1,5 @@
+{{question_text}}
+{% for option in question_options %}
+{{option}} : {{ option_labels.get(option, "") }}
+{% endfor %}
+Only 1 option may be selected.

edsl/questions/templates/list/answering_instructions.jinja

@@ -0,0 +1,4 @@
+Return your answers on one line, in a comma-separated list of your responses, with square brackets and each answer in quotes E.g., ["A", "B", "C"]
+{% if include_comment %}
+After the answers, you can put a comment explaining your choice on the next line.
+{% endif %}

edsl/questions/templates/list/question_presentation.jinja

@@ -0,0 +1,5 @@
+{{question_text}}
+
+{% if max_list_items is not none %}
+The list must not contain more than {{ max_list_items }} items.
+{% endif %}

edsl/questions/templates/multiple_choice/answering_instructions.jinja

@@ -0,0 +1,9 @@
+{# Answering Instructions #}
+{% if use_code %}
+Respond only with the code corresponding to one of the options.
+{% else %}
+Respond only with a string corresponding to one of the options.
+{% endif %}
+{% if include_comment %}
+After the answer, you can put a comment explaining why you chose that option on the next line.
+{% endif %}

edsl/questions/templates/multiple_choice/question_presentation.jinja

@@ -0,0 +1,12 @@
+{# Question Presention #}
+{{question_text}}
+{% if use_code %}
+    {%- for option in question_options %}
+{{ loop.index0 }}: {{option}}
+    {% endfor %}
+{% else %}
+    {% for option in question_options %}
+{{option}}
+    {% endfor %}
+{% endif %}
+Only 1 option may be selected.

edsl/questions/templates/numerical/answering_instructions.jinja

@@ -0,0 +1,8 @@
+This question requires a numerical response in the form of an integer or decimal (e.g., -12, 0, 1, 2, 3.45, ...).
+Respond with just your number on a single line.
+If your response is equivalent to zero, report '0'
+If you cannot determine the answer, report 'None'
+
+{% if include_comment %}
+After the answer, put a comment explaining your choice on the next line.
+{% endif %}

edsl/questions/templates/numerical/question_presentation.jinja

@@ -0,0 +1,7 @@
+{{question_text}}
+{% if min_value is not none %}
+    Minimum answer value: {{min_value}}
+{% endif %}
+{% if max_value is not none %}
+    Maximum answer value: {{max_value}}
+{% endif %}

edsl/questions/templates/rank/answering_instructions.jinja

@@ -0,0 +1,11 @@
+{# Answering Instructions #}
+{% if use_code %}
+Please respond only with a comma-separated list of the code of the raked options, with square brackets. E.g., [0, 1, 3]
+{% else %}
+Please respond only with a comma-separated list of the ranked options, with square brackets. E.g., ['Good', 'Bad', 'Ugly']
+{% endif %}
+{% if include_comment %}
+After the answer, you can put a comment explaining your choice on the next line.
+{% endif %}
+
+

edsl/questions/templates/rank/question_presentation.jinja

@@ -0,0 +1,15 @@
+{{question_text}}
+{% if use_code %}
+The options are
+{% for option in question_options %}
+{{ loop.index0 }}: {{option}}
+{% endfor %}
+{% else %}
+The options are:
+{% for option in question_options %}
+{{option}}
+{% endfor %}
+{% endif %}
+{% if num_selections %}
+You can inlcude up to {{num_selections}} options in your answer.
+{% endif %}

edsl/questions/templates/top_k/answering_instructions.jinja

@@ -0,0 +1,8 @@
+{# Answering Instructions #}
+Please respond with valid JSON, formatted like so:
+{% if include_comment %}
+    {"answer": [<put comma-separated list here>], "comment": "<put explanation here>"}
+{% else %}
+    {"answer": [<put comma-separated list here>]}
+{% endif %}
+

edsl/questions/templates/top_k/question_presentation.jinja

@@ -0,0 +1,22 @@
+{{question_text}}
+{% if use_code %}
+    {% for option in question_options %}
+{{ loop.index0 }}: {{option}}
+    {% endfor %}
+{% else %}
+    {% for option in question_options %}
+{{ option }}
+    {% endfor %}    
+{% endif %}
+
+{# Restrictions #}
+{% if min_selections != None and max_selections != None and min_selections == max_selections %}
+You must select exactly {{min_selections}} options.
+{% elif min_selections != None and max_selections != None %}
+Minimum number of options that must be selected: {{min_selections}}.
+Maximum number of options that must be selected: {{max_selections}}.
+{% elif min_selections != None %}
+Minimum number of options that must be selected: {{min_selections}}.
+{% elif max_selections != None %}
+Maximum number of options that must be selected: {{max_selections}}.
+{% endif %}

edsl/questions/templates/yes_no/answering_instructions.jinja

@@ -0,0 +1,6 @@
+{# Answering Instructions #}
+Please reponse with just your answer. 
+
+{% if include_comment %}
+After the answer, you can put a comment explaining your reponse.
+{% endif %}

edsl/questions/templates/yes_no/question_presentation.jinja

@@ -0,0 +1,12 @@
+{# Question Presention #}
+{{question_text}}
+{% if use_code %}
+    {%- for option in question_options %}
+{{ loop.index0 }}: {{option}}
+    {% endfor %}
+{% else %}
+    {% for option in question_options %}
+{{option}}
+    {% endfor %}
+{% endif %}
+Only 1 option may be selected.

edsl/results/Dataset.py

@@ -8,6 +8,7 @@ from typing import Any, Union, Optional
 import numpy as np
 
 from edsl.results.ResultsExportMixin import ResultsExportMixin
+from edsl.results.DatasetTree import Tree
 
 
 class Dataset(UserList, ResultsExportMixin):
@@ -30,6 +31,15 @@ class Dataset(UserList, ResultsExportMixin):
         _, values = list(self.data[0].items())[0]
         return len(values)
 
+    def keys(self):
+        """Return the keys of the first observation in the dataset.
+
+        >>> d = Dataset([{'a.b':[1,2,3,4]}])
+        >>> d.keys()
+        ['a.b']
+        """
+        return [list(o.keys())[0] for o in self]
+
     def __repr__(self) -> str:
         """Return a string representation of the dataset."""
         return f"Dataset({self.data})"
@@ -245,6 +255,16 @@ class Dataset(UserList, ResultsExportMixin):
 
         return Dataset(new_data)
 
+    @property
+    def tree(self):
+        """Return a tree representation of the dataset.
+
+        >>> d = Dataset([{'a':[1,2,3,4]}, {'b':[4,3,2,1]}])
+        >>> d.tree.print_tree()
+        Tree has not been constructed yet.
+        """
+        return Tree(self)
+
     @classmethod
     def example(self):
         """Return an example dataset.

edsl/results/DatasetExportMixin.py

@@ -4,6 +4,7 @@ import base64
 import csv
 import io
 import html
+from typing import Optional
 
 from typing import Literal, Optional, Union, List
 
@@ -27,6 +28,10 @@ class DatasetExportMixin:
         >>> d.relevant_columns(remove_prefix=True)
         ['b']
 
+        >>> d = Dataset([{'a':[1,2,3,4]}, {'b':[5,6,7,8]}])
+        >>> d.relevant_columns()
+        ['a', 'b']
+
         >>> from edsl.results import Results; Results.example().select('how_feeling', 'how_feeling_yesterday').relevant_columns()
         ['answer.how_feeling', 'answer.how_feeling_yesterday']
 
@@ -37,7 +42,7 @@ class DatasetExportMixin:
         >>> Results.example().relevant_columns(data_type = "flimflam")
         Traceback (most recent call last):
         ...
-        ValueError: No columns found for data type: flimflam. Available data types are: ['agent', 'answer', 'comment', 'model', 'prompt', 'question_options', 'question_text', 'question_type', 'raw_model_response', 'scenario'].
+        ValueError: No columns found for data type: flimflam. Available data types are: ...
         """
         columns = [list(x.keys())[0] for x in self]
         if remove_prefix:
@@ -152,12 +157,13 @@ class DatasetExportMixin:
         iframe_height: int = 200,
         iframe_width: int = 600,
         web=False,
-    ) -> None:
+        return_string: bool = False,
+    ) -> Union[None, str, "Results"]:
         """Print the results in a pretty format.
 
         :param pretty_labels: A dictionary of pretty labels for the columns.
         :param filename: The filename to save the results to.
-        :param format: The format to print the results in. Options are 'rich', 'html', or 'markdown'.
+        :param format: The format to print the results in. Options are 'rich', 'html', 'markdown', or 'latex'.
         :param interactive: Whether to print the results interactively in a Jupyter notebook.
         :param split_at_dot: Whether to split the column names at the last dot w/ a newline.
         :param max_rows: The maximum number of rows to print.
@@ -166,6 +172,9 @@ class DatasetExportMixin:
         :param iframe_height: The height of the iframe.
         :param iframe_width: The width of the iframe.
         :param web: Whether to display the table in a web browser.
+        :param return_string: Whether to return the output as a string instead of printing.
+
+        :return: None if tee is False and return_string is False, the dataset if tee is True, or a string if return_string is True.
 
         Example: Print in rich format at the terminal
 
@@ -249,11 +258,14 @@ class DatasetExportMixin:
 
         >>> r.select('how_feeling').print(format='latex')
         \\begin{tabular}{l}
-        \\toprule
         ...
+        \\end{tabular}
+        <BLANKLINE>
         """
         from IPython.display import HTML, display
         from edsl.utilities.utilities import is_notebook
+        import io
+        import sys
 
         def _determine_format(format):
             if format is None:
@@ -262,7 +274,9 @@ class DatasetExportMixin:
                 else:
                     format = "rich"
             if format not in ["rich", "html", "markdown", "latex"]:
-                raise ValueError("format must be one of 'rich', 'html', or 'markdown'.")
+                raise ValueError(
+                    "format must be one of 'rich', 'html', 'markdown', or 'latex'."
+                )
 
             return format
 
@@ -281,21 +295,24 @@ class DatasetExportMixin:
 
         new_data = list(_create_data())
 
+        # Capture output if return_string is True
+        if return_string:
+            old_stdout = sys.stdout
+            sys.stdout = io.StringIO()
+
+        output = None
+
         if format == "rich":
             from edsl.utilities.interface import print_dataset_with_rich
 
-            print_dataset_with_rich(
+            output = print_dataset_with_rich(
                 new_data, filename=filename, split_at_dot=split_at_dot
             )
-            return self if tee else None
-
-        if format == "markdown":
+        elif format == "markdown":
             from edsl.utilities.interface import print_list_of_dicts_as_markdown_table
 
-            print_list_of_dicts_as_markdown_table(new_data, filename=filename)
-            return self if tee else None
-
-        if format == "latex":
+            output = print_list_of_dicts_as_markdown_table(new_data, filename=filename)
+        elif format == "latex":
             df = self.to_pandas()
             df.columns = [col.replace("_", " ") for col in df.columns]
             latex_string = df.to_latex(index=False)
@@ -305,23 +322,14 @@ class DatasetExportMixin:
                     f.write(latex_string)
             else:
                 print(latex_string)
-
-            return self if tee else None
-
-        if format == "html":
+                output = latex_string
+        elif format == "html":
             from edsl.utilities.interface import print_list_of_dicts_as_html_table
 
             html_source = print_list_of_dicts_as_html_table(
                 new_data, interactive=interactive
             )
 
-            # if download_link:
-            #     from IPython.display import HTML, display
-            #     csv_file = output.getvalue()
-            #     b64 = base64.b64encode(csv_file.encode()).decode()
-            #     download_link = f'<a href="data:file/csv;base64,{b64}" download="my_data.csv">Download CSV file</a>'
-            #     #display(HTML(download_link))
-
             if iframe:
                 iframe = f""""
                 <iframe srcdoc="{ html.escape(html_source) }" style="width: {iframe_width}px; height: {iframe_height}px;"></iframe>
@@ -334,7 +342,18 @@ class DatasetExportMixin:
 
                 view_html(html_source)
 
-            return self if tee else None
+            output = html_source
+
+        # Restore stdout and get captured output if return_string is True
+        if return_string:
+            captured_output = sys.stdout.getvalue()
+            sys.stdout = old_stdout
+            return captured_output or output
+
+        if tee:
+            return self
+
+        return None
 
     def to_csv(
         self,
@@ -453,7 +472,11 @@ class DatasetExportMixin:
         from edsl import ScenarioList, Scenario
 
         list_of_dicts = self.to_dicts(remove_prefix=remove_prefix)
-        return ScenarioList([Scenario(d) for d in list_of_dicts])
+        scenarios = []
+        for d in list_of_dicts:
+            scenarios.append(Scenario(d))
+        return ScenarioList(scenarios)
+        # return ScenarioList([Scenario(d) for d in list_of_dicts])
 
     def to_agent_list(self, remove_prefix: bool = True):
         """Convert the results to a list of dictionaries, one per agent.
@@ -497,7 +520,7 @@ class DatasetExportMixin:
 
         return list_of_dicts
 
-    def to_list(self, flatten=False, remove_none=False) -> list[list]:
+    def to_list(self, flatten=False, remove_none=False, unzipped=False) -> list[list]:
         """Convert the results to a list of lists.
 
         :param flatten: Whether to flatten the list of lists.
@@ -593,7 +616,7 @@ class DatasetExportMixin:
             return filename
 
     def tally(
-        self, *fields: Optional[str], top_n: Optional[int] = None, output="dict"
+        self, *fields: Optional[str], top_n: Optional[int] = None, output="Dataset"
     ) -> Union[dict, "Dataset"]:
         """Tally the values of a field or perform a cross-tab of multiple fields.
 
@@ -601,9 +624,11 @@ class DatasetExportMixin:
 
         >>> from edsl.results import Results
         >>> r = Results.example()
-        >>> r.select('how_feeling').tally('answer.how_feeling')
+        >>> r.select('how_feeling').tally('answer.how_feeling', output = "dict")
         {'OK': 2, 'Great': 1, 'Terrible': 1}
-        >>> r.select('how_feeling', 'period').tally('how_feeling', 'period')
+        >>> r.select('how_feeling').tally('answer.how_feeling', output = "Dataset")
+        Dataset([{'value': ['OK', 'Great', 'Terrible']}, {'count': [2, 1, 1]}])
+        >>> r.select('how_feeling', 'period').tally('how_feeling', 'period', output = "dict")
         {('OK', 'morning'): 1, ('Great', 'afternoon'): 1, ('Terrible', 'morning'): 1, ('OK', 'afternoon'): 1}
         """
         from collections import Counter
@@ -615,6 +640,8 @@ class DatasetExportMixin:
             column.split(".")[-1] for column in self.relevant_columns()
         ]
 
+        # breakpoint()
+
         if not all(
             f in self.relevant_columns() or f in relevant_columns_without_prefix
             for f in fields
@@ -641,6 +668,7 @@ class DatasetExportMixin:
         from edsl.results.Dataset import Dataset
 
         if output == "dict":
+            # why did I do this?
             warnings.warn(
                 textwrap.dedent(
                     """\